Journal of Waterway, Port, Coastal, and Ocean Engineering

Connecting the Mississippi River and the Gulf of Mexico, the Southwest Pass (SWP) is one of the most highly utilized commercial waterways in the United States. Hard-to-predict accumulation of sediments in the SWP affects the access of deep-draft vessels to four of the nation’s top 15 ports measured by tonnage. The U.S. Army Corps of Engineers (USACE) spends approximately 100 Million USD annually on dredging operations to maintain SWP at a 14.2-meter (50-ft.) depth. Presently, USACE project managers rely on rules-of-thumb with seasonal river stage trends and thresholds to get 10–14 days of lead time for shoaling conditions at the SWP. This work presents the development of a machine learning modeling framework to increase lead times and accuracy of shoaling forecasts in the SWP. Within a multivariate multistep timeseries forecasting framework, several regression models, input variables, and forecasting days are explored. All multivariate machine learning models outperformed an univariate ARIMA model used as baseline. A multilayered perceptron regressor implemented on a 60-day in-lag scenario was found to be the best model to forecast shoaling in the upcoming 45 days. The proposed model may be applied to forecast dredging needs at other critical waterways.

Abstract

Among the factors determining the competitiveness of transportation sectors, those associated with travel time play a crucial role. Constant availability of transportation services over time is a key factor. Inland waterway transport is believed to exhibit seasonal fluctuations in cargo flows, implying limited service provision during specific periods in the annual cycle. This article compares the variability of cargo transport over time in inland waterway shipping with that in other land transportation modes. The hypothesis is that the periodic fluctuations in cargo transport during the annual cycle in inland waterway shipping are not significantly different from those in other transport sectors. The analysis of transport variability over time is based on the indicators of periodic fluctuations in land transportation modes from 2005 to 2022. Statistical inference was conducted using variance analysis. As the conditions for applying the ANOVA procedure were not met, the Kruskal–Wallis test was employed as a nonparametric alternative to variance analysis. The research results indicate that there is no basis to reject this hypothesis in the group of examined Western European countries. Different results were obtained in Poland, where cargo transport in inland waterway shipping exhibited more significant variability over time than other transport modes.

Practical Applications

The results obtained from this research can serve as a source of knowledge for entities and institutions responsible for implementing the concept of sustainable development in transportation. The proposed European Green Deal concept in December 2019 states, “The role of rail and inland waterways in land freight transport should be significantly increased first and foremost…”. Despite limited service availability during specific periods of the annual cycle (due to seasonal and random factors), the research results confirm that in countries with well-developed waterways, inland water transport remains competitive compared with other land transport modes. These findings can also be a significant signal for policymakers in countries with extensive inland waterways (e.g., Poland) that may need to fully appreciate the advantages of this mode of transport, partly due to its susceptibility to periodic fluctuations in transport. The research results and formulated conclusions can also be addressed to existing and potential shippers. The confirmed higher or comparable capability of inland navigation to achieve consistent transport volumes during different periods of the annual cycle can be an encouraging factor for incorporating this mode into supply chains.

This paper presents an easy to operate time-dependent, two-dimensional vertical model for the simulation of sand and mud concentrations, transport, and morphodynamic bed evolution in conditions with combined currents and waves. The basic model equation is the advection–diffusion equation for sediments including flow velocities, sediment mixing coefficients, and the settling velocity. The vertical distribution of the sediment mixing coefficient is described by fairly simple and flexible expressions based on flow and wave parameters with user-specified coefficients for calibration of results. The settling velocity is constant or dependent on the sediment concentration to represent mud flocculation and hindered settling processes based on input parameters. The bed boundary condition is modeled by a bed concentration as a function of the bed-shear stress due to currents and waves. The model is valid for sand concentrations and for low and high mud concentrations including fluid mud concentrations. The numerical SUSTIM2DV model has been used to simulate sand and mud transport and bed level changes in dredged (shipping) channels. The model is valid for perpendicular and oblique flow across a dredged channel. The channel slopes should not be too steep (1V:7H or milder), because logarithmic velocity profiles are used. Results of various validation cases of channel deposition of sand, silt, and (fluid) mud in tidal conditions with and without waves are presented. The effect of channel geometry is specifically addressed. The numerical model has also been used to derive sediment trapping factors, which can be used in simpler channel deposition models.

Abstract

The concession of the Peruvian Amazon Waterway (PAW) project seeks to strengthen commercial transport between Brazil and Peru by using four rivers: Amazon, Ucayali, Marañón, and Huallaga, in which the Peruvian Ministry of Transportation and Communications (MTC) was in charge of developing the Terms of Reference (TOR). At the end, the TOR did not specify the methodology for the hydrogeomorphic characterization of rivers. The PAW project was designed to meet specific geometric characteristics to ensure navigable conditions: minimum water depth of 2.44 m (8 ft) and channel width ranging from 44 to 56 m, approximately. The zones where the river bottom limits the layout and dimensions of the navigation channel are called shallow zones (SZs). Based on the PAW project’s TOR, the EIA-d (Detailed Environmental Impact Assessment) study was carried out using 14 SZs that required dredging 3.87 million m³ of sediments. In 2019, the EDI (Definitive Engineering Study) described 24 SZs with a total dredging volume of 4.39 million m³ of sediments. The number and locations of SZs and the volume of required dredging changed significantly, revealing a lack of understanding of meandering (i.e., Ucayali and Huallaga) and anabranching (i.e., Marañón and Amazon) river processes. This study presents an integral methodology for the hydrogeomorphic characterization of the meandering Huallaga River (the river with 56% of the total estimated volume to dredge) by using remote sensing and field measurements (hydrodynamic, sediment transport, and bathymetry). Geological confinement of river dynamics, together with the presence of anabranching structures, promote stable SZs. In regions where meandering rivers tend to develop ancient and modern geomorphic structures, SZs are temporarily and highly dependent of river dynamics. The present hydrogeomorphic characterization of the Huallaga River will support the Peruvian government to inform their technical guidelines (TOR), especially for infrastructure projects (e.g., PAW) in Amazonian rivers.

Practical Applications

Several river-related engineering projects lack a broad and comprehensive view to understand a river’s spatial and temporal dynamics. One of these projects is the Amazon Waterway (Hidrovía Amazónica in Spanish) which sought to strengthen commercial transport between Brazil and Peru using four rivers: Amazon, Ucayali, Marañón, and Huallaga. The Terms of Reference (TOR) for the Environmental Impact Study (EIA) did not specify the methodology for understanding river dynamics; however, the focus was mainly concentrated on the need to develop dredging activities. During the EIA evaluation stage, more than 400 technical observations were presented, several related to fluvial geomorphology. This article presents a synthesis of how proper river studies should be carried out and how river dynamics is related to aquatic biodiversity, highlighting the importance of understanding rivers before developing infrastructure projects.